Table of Contents
Introduction
Confirmation bias connotes the process of interpreting events based on a pre-existing belief or hypothesis (Nickerson, 1998).It is a problematic aspect of human reasoning because it is deemed to be the single most influential perspective that contributes to altercations, misunderstanding, and disputes. However, confirmation bias involves making decisions unintentionally albeit processing information rationally due to an already set opinion or belief. Vaccination has undeniably and significantly reduced the risk associated with the infectious disease the world over. The main infections that have been controlled through vaccination include smallpox, mumps, polio, and measles among others. However, the success of the vaccination programs is almost entirely dependent on the rejection or acceptance of administration by the individuals (Voinson, Billiard, & Alvergne, 2015). Some people believe vaccination is important to prevent the infectious disease but others are opposed due to religious or traditions. It is, therefore, imperative to determine the possibility of overcoming confirmation bias in attitudes towards vaccination using an experimental approach.
In the recent past, measles has reappeared in the United States with a higher magnitude than projected, with more than 640 cases being reported in 2014 (Horne, Powell, Hummerl, & Holyoak, 2015). This has been attributed to the refusal by some parents to vaccinate their children for the fear of adverse side effects. The authors noted that some research studies linked vaccination to autism causing tension among parents across the nation. In another study conducted by Wheeler and Buttenheim (2013), parents increasingly prefer selective vaccination rather than all the vaccinations recommended by the immunization schedule. Most of the narrative regarding adverse effects of vaccination has been influenced by anti-vaccine activists through a series of stories. Once the parents repeatedly hear the same story, they are likely to change their minds and create the illusion that vaccination is evil. In this regard, healthcare providers need to upbeat the pro-vaccine message and encourage parents to let their children have the vaccination.
Attitudes towards vaccination are not only prominent among parents but also among medical students and other people in the society. Busse, Kulkarni, Campbell, and Injeyan (2002) conducted a study to establish the attitudes towards vaccination among Canadian chiropractic students. Students were randomly selected into the study to ensure randomization and then an eleven-item questionnaire administered to them. The questionnaire intended to get students’ perception of vaccines sources among other things. Slightly over 50 percent of the students agreed with vaccination. Moreover, while about 60 percent of first-year students agreed with vaccination, about 40 percent of the fourth year students were opposed to vaccination. The results indicated that as the students progressed in years of education, they were more likely to oppose vaccination. This could have been impacted by exposure to various research studies and beliefs from social settings(Nyhan & Reifler, 2015).
Despite the need to change the belief among some parents on vaccination, it remains a difficult task since their beliefs take precedence of rationale. This is because fluency often leads to errors in judgment and increases the impact of confirmation bias. However, through promoting meta-cognitive difficulty, the parents opposed to vaccination can reduce reliance on beliefs and consider other sources of facts and information (Hernandez and Preston, 2012). Disfluency, therefore, is likely to invoke more reasoning and rationale before taking action as opposed to the common way of thinking. Several studies have been conducted to test the effect of disfluency on confirmation bias. Hernandez and Preston (2012) conducted a research study involving political ideology and positivity towards as court defendant. They concluded that when facts are presented in an unusual way, people tend to more analytical and careful, thereby reducing confirmation bias. However, the research also indicated that while people are under a cognitive load, the conclusion is not necessarily true. As such, confirmation bias can only be overcome through the use of cognitive resources.
Attitudes towards vaccination are also due to anecdotal evidence apart from fluency and disfluency. Anecdotal evidence is based on personal experiences but cannot be confirmed scientifically. According to research conducted by Shelby and Ernst (2013), storytelling can be infectious and lead to various beliefs among people without desire for scientific proof of facts. In the study, the outlined the story of a baby who died of what the parents believed was Hepatitis B vaccination shot. The parents of the deceased took to social media and warned other parents against the administration of the immunization, a call that was supported by numerous sympathizers. Consequently, the information spread across platforms with people questioning the rationale of the claims facing a hostile backlash. Therefore, some parents have developed confirmation bias against vaccines through anecdotal information without due diligence to establish the science behind the claims. The authors suggest that as much as anti-vaccines activism in increasing online, it would be prudent to apply the same anecdotal approach to spread pro-vaccine messages.
In this regard, this research aims at addressing the research question: -In the context of investigating attitudes towards vaccinations, can the confirmation bias be overcome? To answer this question sufficiently, the research was guided by three hypotheses. 1) Disfluent vaccination safety information will result in more positive vaccination attitude change than standard vaccination safety information (presented fluently);2) Anecdotal vaccination safety information will result in more positive vaccination attitude change than standard vaccination safety information, and 3) Disfluent vaccination safety information will result in more positive vaccination attitude change than anecdotal vaccination safety information.
Method
Participants
The respondents in the study were obtained from Psyc105 students, who invited their colleagues and relatives online, participate in the study. The participants were required to fill in an online questionnaire on Qualtrics Survey Software. Hence, a non-probabilistic method was used to obtain a convenience sample. The sample size targeted initially was 1370 participants but due to non-responses encountered, 1069 responses remained. The female participants were 67.20 percent, males 31.87 percent while 0.94 percent formed others category. The participants’ age ranged from 14 to 81 years with the average age of 29.6 years and standard deviation of 14.24.
Materials
Attitude towards vaccination was measured using a study-designed measure scale. The scale tried to emulate Social Desirability scale and hence demonstrated high reliability and validity given the age group under the study. The scale comprised of 11 statements which the respondentwas required to either agree or disagree. The response, agree or disagree, for each statement indicated the respondent’s attitude towards vaccination. Participants were required to respond to the statement before reading the three conditions and after reading conditions. Then, the difference between the second and first vaccination attitude score was obtained.
Procedure
The data collection was done online through Qualtrics Survey Software where participants were required to complete the study. After accepting the terms of the study, they proceeded in answering the questionnaire which was divided into three. The first part was about demographic questions ( gender, age, and education), the second part was answering statements on the designed scale for the study and the last part was fill in the same statements on the scale after reading the three conditions provided. All the participants were invited to participate by psychology students and hence were not allocated to either experimental or control group. After filling in the questionnaire, the questionnaire was supposed to be submitted in the same software for data entry and analysis to take place.
Results
Descriptive Statistics
Based on the results, 33.02 percent of participants were on fluency condition, 31.62 percent on disfluency condition and 35.36 percent on anecdotal condition. More so, most of the respondents were females with 67.20 percent followed by males with 31.87percent. Additionally, slightly less than half of the respondents (49.67 percent) had high school certificates, 24.04 percent had a Bachelor degree, 10.42 percent had a diploma, 7.89 percent and 7.70 percent had a certificate and Masters Degree respectively and the remaining percentage (0.28) had a Ph.D. The mean score for first vaccination attitudes rating scores was; 15.614 for fluency condition with a standard deviation of 1.292, 15.629 for disfluency condition with a standard deviation of 1.237 and mean of 15.460 for the anecdotal condition with a standard deviation of 0.988.
Hypothesis Testing
Independent sample t-test was used to test the hypotheses. Even though the Levene’s test was not conducted, the study assumed equality of variance.
Regarding the first hypothesis, there was a significant difference between fluency condition (M = -0.125, SD = 0.0637) and disfluency condition (M = 1.207, SD = 0.0706); t (689) = -14.0362, p< 0.05. With regard to second hypothesis, there was significant difference between fluency (M= -0.125, SD = 0.0637) and anecdotal condition (M = 1.953, SD= 0.0722); t (729) = -21.4710, p< 0.05. Lastly, in the third hypothesis, there was a significant difference between disfluency condition (M = 1.207, SD = 0.0706) and anecdotal condition (M = 1.953, SD= 0.0722); t (714) = -7.3540, p< 0.05
Discussion
Based on the results obtained, disfluent vaccination safety information resulted in more positive vaccination attitude change than standard vaccination safety information, confirming that the first hypothesis was correct. Similarly, anecdotal vaccination safety information yielded more positive vaccination attitude change than standard vaccination safety information. However, the third hypothesis was rejected since disfluent vaccination safety information resulted in less positive vaccination attitude change than anecdotal vaccination safety information. Therefore, it both anecdotal and disfluency approaches are significantly efficient towards reducing confirmation bias towards vaccination, with the anecdotal intervention being the best alternative. Confirmation bias regarding attitude towards vaccination, therefore, can be reduced in the society by telling stories and sharing information in non-scientific platforms to allow people change their minds through rational and logical reasoning. Additionally, the disfluent approach can also be used for explaining concepts in a non-usual format for the target audience to pay more attention and critically analyze facts about vaccinations devoid of their preexisting beliefs.
This study forms a basis for further research in confirmation bias regarding attitudes towards vaccination. Unlike in the current study, subsequent research should consider a more robust statistical approach such as analysis of variance rather than conducting multiple t-tests. This will reduce the asymptotic errors and have more accurate results. Additionally, it will be prudent to use a random sample as opposed to convenience sampling to incorporate the probabilistic sense into the analysis and enhance validity as well as generalizability of the results.
- Busse, J. W., Kulkarni, A. V., Campbell, J. B., &Injeyan, H. S. (2002). Attitudes toward vaccination: A survey of Canadian chiropractic students. Canadian Medical Association Journal, 166 (12), 1531-1534.
- Hernandez, I., Preston, J. L. (2013). Disfluency disrupts the confirmation bias. Journal of Experimental Social Psychology, 49, 178-182. https://doi.org/10.1016/j.jesp.2012.08.010
- Horne, Z., Powell, D., Hummerl, J. E., &Holyoak, K. K. (2015).Countering antivaccination attitudes.Proceedings of the National Academy of Science, 112, 10321-10324. https://doi.org/10.1073/pnas.1504019112
- Nickerson, R. (1998). Confirmation bias: A ubiquitous phenomenon in many guises. Review of General Psychology, 2, 175-220. https://doi.org/10.1037//1089-2680.2.2.175
- Nyhan, B. &Reifler, J. (2015). Does correcting myths about the flu vaccine work? An experimental evaluation of the effects of corrective information. Vaccine, 33(3):459–464.
- Shelby, A., Ernst, K. (2013). Story and science: How providers and parents can utilize storytelling to combat anti-vaccine misinformation. Human Vaccines &Immunotherapeutics, 9, 1795-1801. Doi: 10.4161/hv.24828.
- Voinson, M., Billiard, S., &Alvergne, A. (2015).Beyond rational decision-making: modeling the influence of cognitive biases on the dynamics of vaccination coverage. PLoS ONE 10(11): e0142990. Doi:10.1371/journal.pone.0142990.
- Wheeler, M.&Buttenheim, A. M. (2013). Parental vaccine concerns, information source, and choice of alternative immunization schedules. Human Vaccines & Immunotherapeutics, 9(8), 1782-1789. Doi:10.4161/hv.25959